Election Data Analyzer

Folks I am interested in us sharing in the comments what each of you is doing to fix the registration roll problems. What is the process after you find a bad registration for fixing it. Secondarily it would be useful to know what features of EDA allow you to make these discoveries. This channel is all about FIXING 2020 not analyzing it to death. Let’s learn from each other. Comments please....

New case study using MapIt step by step. County of 320,000 registrants, I used MapIt and assigned the TYPES for unique address having greater than 8 registrants. Below 8 the TYPES are RESIDENCE. Using FILTER01 tab to extract the non-RESIDENT types from the 320,000 records EXPORT using the EXPORT button. In the export, I inserted the pivot table looking at TYPES group by STATUS and STATUS REASON. In this grouping 61% VOTED. I then broke down each type to see what % VOTED. 37% of the Social Services TYPE VOTED. I then built a second pivot table and chart resulting in the bar chart. There are people voting using the SOCIAL SERVICES address from 2008. Does that make sense? I can understand a short term need (as permitted by law in my state), to use this address type for say the most current election cycles while a person needs those services. But why is the Clerk sending ballots out to this address TYPE for reg dates say older than 2020, for example? Comments?

Now that EDA v6.0 is out, I am going to do some case studies. I took a county of 320,000, ran the EDA calculations and then classified the first 200 address using MapIt. This all took less than 4 hours. I then wanted to focus on looking at Residences TYPES for this case study. That is what the charts are focused on above. Top Left: There are 114 instances of Residences which have 8 registrants, 42 with 9, 13 with 10 etc. Top Right: Of the registrants with the Status - Failed To Vote, UNdeliverable Ballot and Returned Mail, 350+ are Active and 200+ are Inactive. Bottom right, you can see this impacts DEM and REP and Unaffiliated voters. So now the question.... if you are managing a canvass effort for a County, can you use this data to focus your efforts on which doors to knock on first?

Am I the first person to find 8 voters registered at a Funeral Directors Office? Can't make this stuff up. 😔

A county I am examining right now looking at registrations since the 2020 general. What do I see? Steadily increasing numbers with a "pulse" of new registrants about every month. This steady pulse might be from the DMV. I have to ask the clerk. Does the Clerk even know? Why is there a steadily increasing trend? Is this simply a result of people dropping off the rolls? Would you always expect DEM registrations to be above REP registrations without fail for an entire year?

Subsititute the word "Election Data Integrity" for "Coronavirus". TNI was intially an election initiative by the BBC. https://www.bbc.com/mediacentre/latestnews/2020/coronavirus-trusted-news

https://rumble.com/vrpgyw-eda-v6.0-release-training-video.html Wrapping up 2021 with EDAv6.0. You can view the training video here. Seems hard to believe but this EDA channel started back in July with me building something in Excel to satisfy my own curiosity about registration rolls. I had NEVER looked at these data prior to that and I am not an expert. However, clearly there is a need for an easy-to-use Excel based tool for the citizen data analyst that requires no more skills than copy, paste, and click. From 0 to 600+ registered EDA users collaborating on USEIP with respect to election integrity is humbling. More important than that is the need to “Return the Diamonds” in 2022. We need the authentic President back….NOW.

Some homework for EDA users. Pick a major city, run the unique address calculation, classify the top 20 address by TYPE. This is what I found....Look at all of these people using an apartment block address with no unit # assocaied with the address. There are 321 such address at the worst case offender. So the mailman shows up with 321 ballots and decides which box to put them in based on the name? Or do they go back to the Clerk as Undeliverable? Then what? How do address without unit numbers for apartment blocks even make it on the roll? When doing TYPE classification, remember, just because it maps to an apartment block doesn't mean it is OK, it needs a Unit #. Getting closer to EDA v6.0 release.....lots to QA, more than I planned.

One of the research charts that has been missing from EDA has been the ability to correlate the address type to the registration date. For example, there is a large surge of registrations in the weeks leading up to the election. You may want to know: What types of address are these surge registrants coming from? Residences, Prisons, Nursing Homes, Vacant Lots, RV Parks? Well, in EDA 6.0, we are taking the next step to figuring that out. In the coming days, I will announce EDA v6.0 to be available on USEIP with this important feature sure to help our canvassers. I am super excited about it.

Last one for today. In this county, I have only looked so far at unique address that have 9 or more registrants per address and classified those types using the new MapIt feature in EDA. That is 71 address total. It took me less than 2 hours to assign the TYPES. Imagine the time savings compared to driving around the county! Look at the dark blue bars which are "RESIDENCE" types. This is telling you that there are 20 address with 9 registrants and 5 address with 10 registrants...in a normal suburban home. Nothing to see here, right? My guess is the canvassing team might be interested in those address.

Folks, These are CURRENT registration dates going back to 1989 that are associated with a "Greek House" TYPE, aka fraternity or sorority. There should be NO registrants listed before 2016 and I am being leniant in that assessment. Do you really think someone has been living at a Greek House for 5+ years? I think folks are going to like the TYPE feature in EDA, just working on some final QA to ensure it is rock solid before I post it to USEIP. If you are interested in this analysis and you want EDA, now would be a good time to sign up as it takes a few business days to go through the process of being added to USEIP.

Check this out. Leading up to the 2020 General Election, the top two surges in voter registrations by TYPE in this county which has a college town are Fraternity/Sororities and wait for it, Homeless Facilities.

A bit closer to release, just some final fine tuning and QA. This chart is going to show you in stacked column form where the registation are coming from by TYPE per day for a configurable period of time....should be interesting and will be the focus of some case studies after I release it....very soon. This is ficticious data.

One of the research charts that has been missing from EDA has been the ability to correlate the address type to the registration date. For example, there is a large surge of registrations in the weeks leading up to the election. You may want to know: What types of address are these surge registrants coming from? Residences, Prisons, Nursing Homes, Vacant Lots, RV Parks? Well, in EDA 6.0, we are taking the next step to figuring that out. In the coming days, I will announce EDA v6.0 to be available on USEIP with this important feature sure to help our canvassers. I am super excited about it.

EDA v5.4 is now posted on USEIP for your "canvassing pleasure". Now get after it!

Here is the source of this quote. Folks, this speech is an amazing read. https://www.jfklibrary.org/asset-viewer/archives/JFKPOF/034/JFKPOF-034-021

Adding some new dashboard elements to EDA v5.4 which will be released soon, perhaps tomorrow. These are insights into the status of your voter registations for canvassing efforts that can be virtual intially (to focus efforts) and then later, record the actual results. Not real data presented here, just for illustrative purposes. The TYPES and STATUS are user configurable.